Mastercard reveals that advanced AI models have drastically lowered cyberattack costs and increased speed at identifying software vulnerabilities, raising urgent security concerns across global sectors.
Mastercard says the rise of large language models has made it far cheaper to turn a known software flaw into a working cyberattack, cutting the cost to less than $3 in some cases. The card network said the same tools are also helping defenders respond more quickly, but only if organisations improve how they deploy them and tighten their security operations.
The company’s findings come as cyber risk intensifies across sectors that hold valuable data and payment systems. In India, technology, public services and finance together account for about 44 per cent of cyberattacks, according to the report cited by Business Standard, while business systems and customers’ personal and financial information make up more than half of the assets targeted. Mastercard also said frontier models are now almost 90 times faster at identifying zero-day vulnerabilities than earlier generations of models, referring to software flaws that can be exploited before a patch is available.
That speed-up has alarmed security researchers beyond Mastercard. Reuters has previously reported on the growing use of AI to uncover vulnerabilities, while technology publication Tom’s Hardware said Anthropic’s Claude Mythos Preview identified thousands of high-severity zero-day issues across major operating systems and web browsers, prompting a wider industry response. Another analysis described how the window between discovering a flaw and exploiting it has shortened sharply, with some experts warning that long-standing disclosure timelines are no longer enough in an AI-driven threat environment.
Mastercard is also focusing on a newer risk: AI agents, or software that can act and transact on a user’s behalf. The company said firms including Visa, Google and India’s National Payments Corporation are working on protocols for agent discovery, agent-to-agent communication and checkout, but identity and verification remain unresolved. Reshmi Ghosh, Mastercard’s senior vice-president for services, said governance must be introduced carefully because agents can “go rogue”, underscoring the need for slow, tightly controlled roll-outs as companies weigh the benefits of automation against the danger of misuse.
Disclaimer: This article is intended to inform and educate, not to recommend or endorse any financial product, investment or strategy. Please consider your own financial circumstances and seek professional advice where appropriate before making financial decisions.
